RomiChessNG5

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

Mike Sherwin
Posts: 965
Joined: Fri Aug 21, 2020 1:25 am
Location: Planet Earth, Sol system
Full name: Michael J Sherwin

RomiChessNG5

Post by Mike Sherwin »

I stumbled upon post about RomiChessNG5. NG5 was before CCRL started testing RomiChess. I don't remember how I lost the NG5 source code and exe. Maybe it was a hard drive crash. I've had a lot of those over the years. The reason for this post is that it seems NG5 may have been the strongest Romi ever. It tested in independant test to be about +250 elo better than P3j which was very close to P3k CCRL 2375. The test I saw reported NG5 to be 2600+ elo. It would be a shame if NG5 was forever lost. I'd like to know what crazy thing I tried in NG5 that was worth 250 elo. So if anyone has the source code to NG5 or any of the NG series please make it available for download somewhere. If you only have the exe that would be good to have also! :) Thanks.
tmokonen
Posts: 1362
Joined: Sun Mar 12, 2006 6:46 pm
Location: Kelowna
Full name: Tony Mokonen

Re: RomiChessNG5

Post by tmokonen »

Hi Mike,

I looked around for the source code on my hard drives, and so far have not found it, but I did find the executables for the different versions of Romichess NG

http://tonyschess.x10host.com/RomiNG.zip
Mike Sherwin
Posts: 965
Joined: Fri Aug 21, 2020 1:25 am
Location: Planet Earth, Sol system
Full name: Michael J Sherwin

Re: RomiChessNG5

Post by Mike Sherwin »

tmokonen wrote: Tue Oct 15, 2024 7:57 pm Hi Mike,

I looked around for the source code on my hard drives, and so far have not found it, but I did find the executables for the different versions of Romichess NG

http://tonyschess.x10host.com/RomiNG.zip
Hi Tony, Wow! Thanks a bunch. :D

TalkChess is really slow right now and the link isn't working. Maybe too many people are downloading it at the moment. If Graham Banks is interested in testing RomiChessNG5 I hope he can get a copy. That would be incredible after all this time! :)
Mike Sherwin
Posts: 965
Joined: Fri Aug 21, 2020 1:25 am
Location: Planet Earth, Sol system
Full name: Michael J Sherwin

Re: RomiChessNG5

Post by Mike Sherwin »

tmokonen wrote: Tue Oct 15, 2024 7:57 pm Hi Mike,

I looked around for the source code on my hard drives, and so far have not found it, but I did find the executables for the different versions of Romichess NG

http://tonyschess.x10host.com/RomiNG.zip
I got it. First I went to http://tonyschess.x10host.com then I downloaded TonyChess003.zip. I'll be taking a look. :) Then I added /RomiNG.zip to the end of the partial URL and it worked. Thanks again! :D
EDIT: Sorry I have to take that back (not the thanks or the smiles). I saw it download but it is not on my hard drive.
EDIT2: Okay I did get it, sort of. Something I've never seen before. I had to send it to Google for scanning before it would show up in my download file. But I have it now, promise! :D
tmokonen
Posts: 1362
Joined: Sun Mar 12, 2006 6:46 pm
Location: Kelowna
Full name: Tony Mokonen

Re: RomiChessNG5

Post by tmokonen »

Mike Sherwin wrote: Tue Oct 15, 2024 9:19 pm I got it. First I went to http://tonyschess.x10host.com then I downloaded TonyChess003.zip. I'll be taking a look. :) Then I added /RomiNG.zip to the end of the partial URL and it worked. Thanks again! :D
EDIT: Sorry I have to take that back (not the thanks or the smiles). I saw it download but it is not on my hard drive.
My hosting has been acting strangely today. I tried uploading to my mega.nz account. Maybe you will have better luck with that.

https://mega.nz/file/zkkzGKSJ#H3bb-MZK7 ... FduEKIBzDU

Edit: I see you got it anyway. All's well that ends well.
Mike Sherwin
Posts: 965
Joined: Fri Aug 21, 2020 1:25 am
Location: Planet Earth, Sol system
Full name: Michael J Sherwin

Re: RomiChessNG5

Post by Mike Sherwin »

RomiChess64P3n - Gnuchessx : 10.5/14 9-2-3 (1=011=1101111= ) 75% +191
RomiChessNG5 - Gnuchessx : 13.0/14 12-0-2 (11=1111=111111 ) 93% +449

It is early, yet first test is around +250 elo over P3n! :D

There is one little teeny tiny problem though.

RomiChess64P3n - RomiChessNG5 : 12.0/15 10-1-4 (=011=1=1=111111) 80% +241

How? :?
Guenther
Posts: 4718
Joined: Wed Oct 01, 2008 6:33 am
Location: Regensburg, Germany
Full name: Guenther Simon

Re: RomiChessNG5

Post by Guenther »

Mike Sherwin wrote: Wed Oct 16, 2024 2:02 am RomiChess64P3n - Gnuchessx : 10.5/14 9-2-3 (1=011=1101111= ) 75% +191
RomiChessNG5 - Gnuchessx : 13.0/14 12-0-2 (11=1111=111111 ) 93% +449

It is early, yet first test is around +250 elo over P3n! :D

There is one little teeny tiny problem though.

RomiChess64P3n - RomiChessNG5 : 12.0/15 10-1-4 (=011=1=1=111111) 80% +241

How? :?
Extremely small samples - and likely flawed testing conditions... well, you asked ;)

P.S. I found some sources of older Romichess versions but not for NG5.
https://rwbc-chess.de

[Trolls n'existent pas...]
Mike Sherwin
Posts: 965
Joined: Fri Aug 21, 2020 1:25 am
Location: Planet Earth, Sol system
Full name: Michael J Sherwin

Re: RomiChessNG5

Post by Mike Sherwin »

Guenther wrote: Wed Oct 16, 2024 8:57 am
Mike Sherwin wrote: Wed Oct 16, 2024 2:02 am RomiChess64P3n - Gnuchessx : 10.5/14 9-2-3 (1=011=1101111= ) 75% +191
RomiChessNG5 - Gnuchessx : 13.0/14 12-0-2 (11=1111=111111 ) 93% +449

It is early, yet first test is around +250 elo over P3n! :D

There is one little teeny tiny problem though.

RomiChess64P3n - RomiChessNG5 : 12.0/15 10-1-4 (=011=1=1=111111) 80% +241

How? :?
Extremely small samples - and likely flawed testing conditions... well, you asked ;)

P.S. I found some sources of older Romichess versions but not for NG5.
Those other testers probably didn't test right either because NG5 really flopped hard after my last post. No need to even talk about it anymore. Thanks for looking! :)
User avatar
Brunetti
Posts: 424
Joined: Tue Dec 08, 2009 1:37 pm
Location: Milan, Italy
Full name: Alex Brunetti

Re: RomiChessNG5

Post by Brunetti »

Mike Sherwin wrote: Wed Oct 16, 2024 10:54 am Those other testers probably didn't test right either because NG5 really flopped hard after my last post. No need to even talk about it anymore. Thanks for looking! :)
I launched a RR tournament yesterday, not yet finished, and NG5 is not performing well vs. its brothers:

Code: Select all

1: RomiChess P3nX64    226.5/363
2: RomiChess P3L       203.0/363
3: RomiChess P3k 64 ja 186.0/364
4: RomiChess P3jx64    149.0/364
5: RomiChess NG5       144.5/364
Alex
User avatar
Brunetti
Posts: 424
Joined: Tue Dec 08, 2009 1:37 pm
Location: Milan, Italy
Full name: Alex Brunetti

Re: RomiChessNG5

Post by Brunetti »

Tournament finished:

Code: Select all

Rank Name                  Elo    +    - games score oppo. draws
   1 RomiChess P3nX64       71   28   28   400   63%   -18   36%
   2 RomiChess P3L          30   27   27   400   55%    -7   37%
   3 RomiChess P3k 64 ja     9   28   28   400   51%    -2   34%
   4 RomiChess P3jx64      -55   28   28   400   40%    14   28%
   5 RomiChess NG5         -55   28   28   400   40%    14   31%
Alex